339 - CO2 Monitoring in Extremely Low Birth Weight Premature Infants- Comparison Between Transcutaneous and blood gas CO2

Sunday, April 30, 2023

3:30 PM – 6:00 PM ET

Poster Number: 339
Publication Number: 339.345

Liron Borenstein-Levin, Rambam Medical center, Haifa, Hefa, Israel; Noa Avishay, Rambam Medical Center, Rehovot, HaMerkaz, Israel; Orit Soffer, Rambam Medical Center, Haifa, Hefa, Israel; Shmuel Arnon, Kupat Cholim, Kfar-Saba, HaMerkaz, Israel; Arieh Riskin, Technion Israel Institute of Technology, Haifa, Hefa, Israel; Gil Dinur, Rambam Medical Center, haifa, Hefa, Israel; Amir Kugelman, Rambam Medical Center, Faculty Of Medicine, Technion, Haifa, Israel, Haifa, Hefa, Israel; Ori Hochwald, Rambam Medical Center, Haifa, Israel, Haifa, Hefa, Israel

Presenting Author(s)

LB

Liron Borenstein-Levin, MD (she/her/hers)

Neonatolgist
Rambam Medical center
Haifa, Hefa, Israel

Background: Premature infants are susceptible to hyper- or hypocapnia and rapid fluctuations in pCO₂, especially during the first week of life. Transcutaneous CO₂ (TcCO₂) monitor allows continuous non-invasive CO₂ monitoring during invasive ventilation, including high frequency oscillatory ventilation (HFOV) as well as during non-invasive ventilation and is becoming more popular in the NICU.

Objective:

To evaluate the agreement between CO₂ levels measured by T_CCO₂monitor and blood gas CO₂ (bgCO₂), among extremely low birth weight (ELBW) premature infants, during the first week of life.

Design/Methods:

A prospective, observational, multicenter study. All premature infants< 1000g admitted to the participating NICUs during the study period were monitored by T_CCO₂ monitor (Sentec, Therwil, Switzerland), if available, during the first week of life. For each bgCO₂ measurement, a simultaneous TcCO₂ measure was documented, as well as demographic and clinical data.

Results:

1603 pairs of T_CCO₂-bgCO₂ values of 85 infants were collected. Median (IQR) birth weight 802 (696-915) g, gestational age 26.4 (26.0-28.3) weeks. A good agreement (Bias±SD: 3.6±9.2 mmHg) and correlation (Pearson r=0.63, p< 0.001) were found between the TcCO₂ and bgCO₂ values (Figure 1, a and b). In 58% of the paired samples the TcCO₂ reading was within ±5 mmHg range as compared to bgCO₂, and in 80% of samples the absolute difference was ≤10 mmHg. Multivariable regression analysis revealed that sampling during the first 3 days of life and venous sampling were independent factors associated with TcCO₂-bgCO₂ absolute difference ≤5 mmHg, and that HFOV increased the likelihood of an absolute TcCO₂-bgCO₂ difference of >5 mmHg. During the first 3 days of life, a moderate correlation was found between the trending of each two successive measures of TcCO₂ vs. bgCO₂- r=0.55. Examples of TcCO₂ and bgCO₂ correlation and trends for two participants are shown in figure 2.

Conclusion(s):

CO₂ measures by TcCO₂ are in good agreement and correlation with bgCO₂ among ELBW premature infants during the first days of life. The bias and mainly the precision of TcCO₂ dictate the use of this continuous non-invasive method as a complimentary tool along with blood gas sampling in order to evaluate CO₂ levels and trending in individual patients.